summaryrefslogtreecommitdiff
path: root/npf.c
AgeCommit message (Collapse)Author
2005-12-08(svn r3271) Remove some debug statements of mine which somehow got into svn.hackykid
2005-11-05(svn r3142) Remove a now invalid case fall-through and use helper function ↵peter1138
instead of map access.
2005-10-05(svn r3019) -Codechange: Replace explicit shifting/anding/oring with GB and SBtron
2005-09-11(svn r2942) Staticise some functionstron
2005-09-09(svn r2929) * Move DistanceTrack from map.c to npf.c and rename to ↵matthijs
NPFDistanceTrack. * Make NPFDistanceTrack return the distance multiplied by NPF_TILE_LENGTH to prevent rounding This should make ship and train pathfinding more accurate and faster. * Update IsEndOfLine to prevent trains from trying to go off a slope onto a tunnel entrance.
2005-07-24(svn r2701) Insert Id tags into all source filestron
2005-07-23(svn r2696) Remove stray semicolonstron
2005-07-22(svn r2674) - CodeChange: [pbs] Generalise the PSBISPbsDepot function so it ↵hackykid
can check if an arbitrary junction is a pbs junction. Preparations for making pbs more safe.
2005-07-22(svn r2673) Include functions.h directly, not globally via openttd.htron
2005-07-17(svn r2625) - Fix: [pbs] Store the end of a train's reserved path ↵hackykid
explicitly. Prevents trains from unreserving eachothers paths in some cases. - CodeChange: Use the TrackdirToTrack function instead of &7, and remove an unneeded variable.
2005-07-15(svn r2585) - Fix [Makefile]: some small cleanups, remove warnings, and add ↵Darkvater
mersenne to makefile (Luca)
2005-07-13(svn r2558) Change the internal map format from 7 arrays to one array of ↵tron
structs, this doesn't change the saved format for now. It's a stepping stone for further changes.
2005-07-09(svn r2538) - Fix: Use IsCompatibleRailType() function instead of checking ↵hackykid
this yourself.
2005-07-04(svn r2517) - Fix: [pbs] Detect end-of-lines properly regarding depots. hackykid
2005-07-04(svn r2516) - Feature: [pbs] Implement path-based-signalling. This allows ↵hackykid
multiple trains within the same signal block, provided their paths dont intersect. For this the block must have all exit and entry signals be pbs signals. Place these by ctrl-clicking 4 times on a normal signal. - Feature: [pbs] Implement autoplacement of pbs blocks, when a block has an entry and an exit pbs signal, covert the entire block to pbs. Can be turned off in the patch settings. - Feature: [pbs] Allow showing of reserved status by making the tracks darker, when the pbs debug level is at least 1.
2005-07-03(svn r2514) - Codechange: [NPF] Move the checking of railtype into a ↵matthijs
funciton IsCompatibleRail(). - Codechange: [NPF] Check the railtype along a route against the engine type instead of against the previouse tile. This clears the way for electriefied rails. - Add: [NPF] [ 1209644 ] A penalty for crossings (peter1138)
2005-06-25(svn r2489) static, bracing style and use clamp()tron
2005-06-25(svn r2487) Replace TILE_XY by TileXY/TileDiffXYtron
2005-06-22(svn r2473) - Add: VehicleMayEnterTile(), which checks if the tile owner of ↵matthijs
a tile is correct for a vehicle to enter it. Based upon glx's code. - Fix: [ 1203769 ] [NPF] NPF tries to plan over bridges, through tunnels, over level crossings of other players. (glx) - Codechange: Renamed TRANSPORT_MAX to TRANSPORT_END and added INVALID_TRANSPORT. - Codechange: Moved IsLevelCrossing() from tile.h to rail.h - Add: GetCrossingTransportType(), which returns the transport type (road, rail) of both tracks on a level crossing. - Removed old TODO that was fulfilled already.
2005-06-22(svn r2471) - Fix: [ 1221249 ] [NPF] Vehicles try to drive into a tunnel ↵matthijs
entrance from above.
2005-06-17(svn r2456) * Prettyfied npf.c using enums and wrappers from rail.h.matthijs
* Moved GetTileRailType() from npf.c to rail.[ch].
2005-06-17(svn r2451) * Fix: Assertion caused by passing a trackdir where a track was ↵matthijs
expected.
2005-06-17(svn r2450) * Codechange: Replaced all uses of the arrays in tile.h with ↵matthijs
calls to the associated wrapper functions. * Codechange: Made npf.c use some map array accessing wrappers instead of direct access. * Codechange/Fix: Named every enum in tile.h. Fixes a nasty bug on MSVC where arrays would be initialised with zeroes (tnx Asterix_) * Removed magic numbers from tables in tile.c. * Added some explicit casts in tile.h.
2005-06-16(svn r2448) General cleanup of rail related code, more to follow.matthijs
* Add: rail.[ch] for rail-related enums and wrapper functions. * Codechange: Removed dozens of magic numbers with below enums. * Codechange: Rewrote CheckTrackCombination(). * Add: TILE_SIZE, TILE_PIXELS and TILE_HEIGHT constants. * Add: enums RailTileType, RailTileSubtype, SignalType to mask against the map arrays. * Add: enums Track, TrackBits, Trackdir, TrackdirBits for railway track data. (Note that the old RAIL_BIT constants are replaced by TRACK_BIT ones). * Add: enums Direction and DiagDirection * Codechange: Moved a bunch of track(dir) related lookup arrays from npf.[ch] to rail.[ch]. * Codechange: move RailType enum from tile.h to rail.h. * Add: Wrapper functions for masking signal status in the map arrays: SignalAlongTrackdir, SignalAgainstTrackdir and SignalOnTrack. * Add: Wrapper functions to access rail tiles, using above enums * Add: Wrapper functions to modify tracks, trackdirs, directions, etc. * Add: Wrapper functions for all lookup arrays in rail.[ch] (Arrays are still used in parts of the code) * Codechange: Renamed some variables and arguments to better represent what they contain (railbit -> track, bits -> trackdirbits, etc.). * Codechange: Don't use FindLandscapeHeight() in CmdRemoveSingleRail(), since it returns way too much info. Use GetTileSlope() instead. * Codechange: [NPF] Removed some unused globals and code from npf.c.
2005-06-16(svn r2447) * Add: [NPF] Penalty for road vehicles making turns.matthijs
2005-06-07(svn r2433) - CodeChange: unmagicify all road/train crossings with ↵Darkvater
IsLevelCrossing() function (peter1138)
2005-06-02(svn r2397) - CodeChange: rename all "ttd" files to "openttd" files.Darkvater
2005-05-23(svn r2363) - Codechange: [NPF] Removed caching the endnode check. The code ↵matthijs
for this was buggy, fundamentally flawed and barely useful. (Hackykid)
2005-05-07(svn r2281) - Fix: [ 1115204 ] [NPF] When pressing the goto depot button, ↵matthijs
trains will now also look behind it if there is no depot in front. If so, the train reverses immediately. This also work anywhere, not just at stations. - Add: [NPF] Reversing inside of depots now has a penalty. It also applies to trains only, other vehicles shouldn't bother reversing. - Fix: [NPF] When checking whether to reverse a train, the trackdir of the first loc was used instead of the last vehicle as a starting node for pathfindig. This might have caused some trains not reversing when they should have (or vice versa). Typo introduced when converting to GetVehicleTrackdir() in r2256. - CodeChange: [NPF] Removed duplicate code by letting NPFRouteTjoStationOrTile() call NPFRouteToStationOrTileTwoWay(). - Add: [NPF] NPFRouteToDepotBreadthFirstTwoWay() to find a depot while also looking backwards. - Add: It is now possibly to specify a path cost for aystar starting nodes.
2005-05-03(svn r2257) - Fix: [NPF] NPF debug markings modify _map2 instead of _map3_hi ↵matthijs
for street tiles, corrupting them. Some info got moved around in r1768 for street tiles, but NPF did not get updated.
2005-05-02(svn r2255) - Fix: [ 9680363 ] [NPF] Broken buoy handling for shipsmatthijs
Buoys will now try to get within 3 tiles of a buoy instead of a the actual buoy tile. This gets ships to got past buoys in a realistic (IMO) way instead of barging right through them. - Fix: [NPF] Trains get curves penalties sometimes even when the track is straight. - Add: [NPF] Ships get a penalty for going over buoys now, so they will try to go around. - Add: [NPF] Ships get a penalty for curves too, yay for straight lines. - Add: TrackdirToTrack(), TrackToTrackdir(), IsDiagonalTrack() and IsDiagonalTrackdir() helper functions. - Add: IsBuoy() and IsBuoyTile() helper functions. - Codechange: Rearranged part of the control flow of ShipController(), removing a goto.
2005-05-02(svn r2253) - Fix: [ 1190896 1184378 ] [NPF] Trains ignoring their railtype ↵matthijs
(mono, maglev) (glx)
2005-04-15(svn r2204) - Add: [NPF] NPF now has a maximum number of nodes it will ↵matthijs
search. The default value is 5000 for now, which is an educated guess. Probably needs some finetuning. Hopefully this "feature" can be removed later on, when more sophisticated means of limiting the pathfinder have been implemented. This should make ships and larger networks playable for now, though.
2005-04-11(svn r2182) - Add: [NPF] There is now a debug class for NPF. Use -d ↵matthijs
npf<level> to enable debugging printouts from npf. - Codechange: [NPF] Removed NPF_MARKROUTE macro, to mark routes just specify a npf debugging level >= 1 on the commandline.
2005-04-11(svn r2181) - Add: DistanceTrack() to calculate the distance over optimally ↵matthijs
laid out tracks. - Codechange: [NPF] Removed unused heuristic function NPFCalcTileHeuristic(). - Codechange: [NPF] Use DistanceTrack() instead of DistanceManhattan() for ship and train heuristic. - Codechange: Renamed variables x and y to dx and dy in some of the distance calculation functions.
2005-04-07(svn r2165) - Codechange: [NPF] Properly enummed NPF hash size, it is easily ↵matthijs
changable now. - Codechange: [NPF] Improved the NPF hash calculation slightly. - Codechange: [NPF] Increased hash size, should speed up somewhat.
2005-04-06(svn r2159) - Fix: [NPF] Road vehicles never found their target station or ↵matthijs
depots (introduced in r2154) - Fix: [NPF] Trains still tried to go through the back of depots.
2005-04-05(svn r2154) - Fix: [NPF] Vehicles should no longer try to drive through the ↵matthijs
back of depots and road stations. - Add: GetDepotDirection() wrapper function. - Fix: [NPF] Ships can now actually reach buoys.
2005-04-04(svn r2148) - Add: [NPF] Trains can now plan taking into account that they ↵matthijs
can reverse in depots. This makes forced servicing tracks work with NPF.
2005-04-04(svn r2147) - Add: [NPF] Give red presignal exit signals a different ↵matthijs
(higher) penalty, to discourage trains from waiting at presignal exits.
2005-04-02(svn r2121) -Fix: changed the 2nd param of AyStar_EndNodeCheck back to what ↵truelight
it should be
2005-03-14(svn r2007) - Fix: [NPF] Slope penalties did not work correctly with ↵matthijs
foundations. (HackyKid) - Fix: [NPF] Stations penalties were not applied correctly, since stations had no base cost. (HackyKid) - Fix: [NPF] Lastred penalty was applied multiple times for every red signal, instead of just the last one. (HackyKid)
2005-03-13(svn r2006) - Fix: [NPF] Wrong signal detection for last signal red ↵matthijs
detection. Multiple tracks per tile with only one signal were detected wrong. (HackyKid)
2005-03-08(svn r1968) - Fix: [NPF] Mixed declarations and statementsmatthijs
2005-03-08(svn r1967) Codechange: A mix of mostly indentation-related tidyups.pasky
2005-03-08(svn r1964) - Add: [NPF] Added a penalty matthijs
2005-03-08(svn r1963) - Add: [NPF] Penalty for a red signal that is the last signal on ↵matthijs
the path. - Add: [NPF] NPFGetFlag() and NPFSetFlag() to wrap NPF node flag handling
2005-03-08(svn r1957) Compilation fix.pasky
2005-03-07(svn r1956) -Fix: [NPF] New target tile for heuristic should perform better ↵matthijs
with larger stations (HackyKid)
2005-02-07(svn r1843) - Codechange: [NPF] Removed some unused code.matthijs